Case Study

CardiAI uses AI to detect the signs of heart disease

Dr Anmol Kapoor has seen countless patients pass through his office throughout his medical career. However, it was one man in particular that changed the Calgary-based cardiologist’s perspective on healthcare, and eventually lead him to explore using AI to help patients with heart disease.

The patient was suffering from heart failure. As his condition worsened, his lungs began to fill with fluid, and breathing was difficult. Just coming into Kapoor's office was a struggle. The doctor would then have to order tests, which meant the man would need to go to the lab and then sometimes back to the office. And then repeat.

"The time came when he said, 'look, I know you really want to help me. You really want to keep me alive. But I'm exhausted. I want to go. Please let me go,'" Kapoor recalls.

Eventually, the patient passed away. However, the memory of his tiring march through countless tests and treatments stuck with Kapoor. It inspired him to develop ways to use technology to improve the health care experience for his patients.

"We should not be practising medicine this way when we have access to technology and have access to the next generation innovations. And we should not be repeating mistakes and doing the same things again and again. We need to innovate and leverage technological advances. We need to find better ways to provide precision health to our patients in a timely manner. "

In 2018, Kapoor started CardiAI, a biotechnology company developing new devices and methods to improve healthcare efficiency. The company's reach is wide; aside from running medical imaging clinics across Alberta, CardiAI participates in clinical studies, builds medical devices and does research into mobile health care. Its subsidiary, BioAro is also developing AI-based genomic testing solutions that will soon be launched in Canada.

Kapoor and his company are exploring projects that use machine learning to improve patient care in cardiology. CardiAI took part in the Prairies Canada Regional Innovation Ecosystems (RIE) program, which supports small companies finding new uses for artificial intelligence in healthcare.

Searching for subtle clues

Support from the program allowed Kapoor to pursue two heart-related projects. One would train a machine learning system to scour through data from Holter monitors. The device is a common tool used to diagnose potential heart problems, consisting of a small box connected to electrodes attached to the patient's chest. The Holter records every heartbeat, generally for 24 to 48 hours – although it can be worn as long as two weeks. A cardiologist then analyzes the data to look for potential signs of arrhythmias and other abnormalities. The machine learning system would go through the data produced by the monitor, looking for anomalies that may indicate a problem.

The second project involves using artificial intelligence to examine images taken of the heart in the field of nuclear medicine. It will help improve the sensitivity and specificity of myocardial perfusion imaging. The system will also get trained on CT scans, coronary angiograms, echocardiograms, or other images of healthy and damaged hearts. Once properly trained, it could then examine new cardiac images and spot warning signs of a patient who needs cardiac care in a timely manner.

While both the Holter monitor and cardiac image projects use very different data, the goal is the same: developing AI systems that can detect the signs of heart disease more accurately and sooner than our current methods. He hopes machine learning tools might be able to pick up on subtle anomalies in heart data that could be missed by cardiologists, especially early on. Even more so, an algorithm might be able to make connections that a person never could. Kapoor says there are so many different factors that can contribute to the health of a heart, more than any one person could take into account. However, that’s a task that machine learning excels at.

We want to close the huge healthcare gap we have. With AI, we can give better care to large segments of the population, reduce economic and healthcare disparity, and improve access.

Dr Anmol Kapoor

Making better predictions

Given how many people are diagnosed with heart disease each year, even small successes could have a big impact. If a machine learning system could assist cardiologists in being even five per cent more accurate when diagnosing a patient, Kapoor says, it would mean many more lives would be saved.

Kapoor says he didn't know much about artificial intelligence when he first started working in cardiology. But as he learned more about it, he saw it as a natural pairing. Even more than most medical disciplines, cardiology is about trying to predict the future.

"Who is at high risk of death, who is at high risk of a heart attack? It's all predictive analysis," he says.

"AI helps with that predictive analysis."

Using machine learning to analyze cardiac images offers another benefit: it allows cardiologists to focus on the cases that need their attention. Kapoor notes that it can take hours for a doctor to comb through a patient's cardiac data, even if they turn out to be healthy. In any healthcare system, there are only so many resources available. And there are already concerns about the number of cardiologists working compared to the patients they see. Artificial intelligence could significantly cut the time physicians spend analyzing that data by flagging the cases that need urgent attention from a human doctor. This assistance allows the cardiologists to make better use of their limited time.

"We want to close the huge healthcare gap we have. With AI, we can give better care to large segments of the population, reduce economic and healthcare disparity, and improve access," Kapoor says.

So far, CardiAI is training its system on data from both healthy and damaged hearts. Initial results have shown that the approach works, he says. They are now focusing on adding new data points to increase the accuracy of the system's predictions.


This project was part of the Western Economic Diversification Canada Regional Innovation Ecosystems (RIE) program. The initiative brought together nine organizations from non-profit, business and academia to establish viable uses for artificial intelligence and machine learning in health and data analytics.

Learn more about how artificial intelligence and machine intelligence can transform your business. Discover 5 ways AI can be your business’ competitive advantage.


Let's get started!

Interested in learning more about how Amii can help you harness the potential of AI in your business?

The first step is easy: contact us! We're here if you want to float an idea or to answer any burning questions you might have. We're always happy to chat with a fellow AI enthusiast (it's absolutely our favourite topic of discussion).

Related Articles

Connect with the community

Get involved in Alberta's growing AI ecosystem! Speaker, sponsorship, and letter of support requests welcome.

Explore training and advanced education

Curious about study options under one of our researchers? Want more information on training opportunities?

Harness the potential of artificial intelligence

Let us know about your goals and challenges for AI adoption in your business. Our Investments & Partnerships team will be in touch shortly!